Automatically expanding segments of user embeddings using multiple user embedding representation types

A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computing device to perform operations comprising:

• generating user embeddings for a plurality of users generated from structured user data utilizing a neural network trained to create user embeddings that encode user profile data into uniform user embeddings;
• determining, from a first digital content campaign, a base segment of users from the plurality of users based on user-provided parameters;
• plotting, within a graphical user interface, user embeddings corresponding to the base segment of users;
• receiving, via one or more selectable graphical elements within the graphical user interface, a selection of a user embeddings type and a user similarity metric;
• identifying one or more clusters of additional user embeddings from the user embeddings for the plurality of users generated utilizing the neural network, the one or more clusters of additional user embeddings having the selected user embeddings type that satisfy the user similarity metric based on distances between the one or more clusters of additional user embeddings and one or more user embeddings of the segment of user embeddings, the one or more clusters of additional user embeddings representing an additional segment of users from the plurality of users;
• generating a visualization of an expanded segment of users comprising the user embeddings corresponding to the base segment of users and the one or more clusters of additional user embeddings;
• and
• receiving a request to target, in an additional digital content campaign, the expanded segment of users.

A system for visualizing user embeddings in low-dimensional space comprising:

• a memory device that comprises:
• a neural network trained to generate the user embeddings from user profile data;
• and
• at least one processor configured to cause the system to:
• generate, utilizing the neural network, a plurality of user embeddings for a plurality of users from the user profile data, the plurality of user embeddings comprising a plurality of user embeddings types;
• receiving, via one or more selectable graphical elements within a graphical user interface, a selection of a user embeddings type and a user similarity metric;
• provide, in response to receiving the selection of the user embeddings type and the user similarity metric within the graphical user interface, a visualization that displays a segment of user embeddings of the selected user embeddings type from the plurality of user embeddings, the segment of user embeddings representing a base segment of users of the plurality of users from a first digital content campaign;
• expand, in response to user interaction with the one or more selectable graphical elements within the graphical user interface and for an additional digital content campaign, the segment of user embeddings by:
• determining one or more clusters of additional user embeddings from the plurality of user embeddings generated using the neural network that are similar to one or more user embeddings of the segment of user embeddings based on at least one of the selected user embeddings type or distances between the one or more clusters of additional user embeddings and user embeddings of the segment of user embeddings, the one or more clusters of additional user embeddings representing an additional segment of users of the plurality of users, and
• adding the one or more clusters of additional user embeddings to the segment of user embeddings;
• provide, within the graphical user interface, an updated visualization of the expanded segment of user embeddings;
• and
• receive a request to target, in the additional digital content campaign, an expanded segment of users corresponding to the expanded segment of user embeddings.